Newsletter Performance Review: 8 sends, Aug 4 to Sep 22, 2026 (fictional data)
Goal: sponsor revenue, with sponsors paying per click. So clicks per send and click rate are the metrics that matter most here. Open rate is shown for completeness only: privacy features such as Apple Mail Privacy Protection preload emails and record "opens" that no human made, so opens are inflated and not a reliable success measure.
Data check: 8 sends, which meets the minimum, but 8 is still a small sample, so the patterns below are directional.
Assumption: your file does not say whether opens and clicks are unique, so I treated them as unique opens and unique clicks. Click rate = clicks / delivered; click to open rate = clicks / opens; unsubscribe rate = unsubscribes / delivered.
Optional inputs not provided: per link click counts (so I could not rank content types by link or measure sponsor link clicks directly) and send traits such as length and lead story (so those traits are not testable).
Summary table per send
| Date | Subject | Delivered | Clicks | Click rate (on delivered) | Click to open rate | Unsub rate | Open rate (unreliable) |
|---|---|---|---|---|---|---|---|
| Aug 4 | 5 tools we cut | 10,200 | 410 | 4.02% | 8.04% | 0.118% | 50.0% |
| Aug 11 | The pricing page teardown | 10,300 | 720 | 6.99% | 13.85% | 0.087% | 50.5% |
| Aug 18 | Weekly roundup #31 | 10,350 | 250 | 2.42% | 5.10% | 0.145% | 47.3% |
| Aug 25 | Weekly roundup #32 | 10,400 | 240 | 2.31% | 5.00% | 0.173% | 46.2% |
| Sep 1 | How Acme doubled trials | 10,500 | 690 | 6.57% | 12.32% | 0.076% | 53.3% |
| Sep 8 | Weekly roundup #33 | 10,520 | 230 | 2.19% | 4.89% | 0.190% | 44.7% |
| Sep 15 | 3 onboarding emails to steal | 10,600 | 650 | 6.13% | 12.04% | 0.094% | 50.9% |
| Sep 22 | Weekly roundup #34 | 10,650 | 220 | 2.07% | 4.78% | 0.207% | 43.2% |
Medians and outliers
| Metric | Median | Range |
|---|---|---|
| Click rate on delivered | 3.22% | 2.07% to 6.99% |
| Click to open rate | 6.57% | 4.78% to 13.85% |
| Unsubscribe rate | 0.131% | 0.076% to 0.207% |
Outlier rule: click rate above 1.5x the median (above 4.83%) or below 0.5x the median (below 1.61%).
| Send | Click rate | x median | Outlier |
|---|---|---|---|
| The pricing page teardown (Aug 11) | 6.99% | 2.17x | High |
| How Acme doubled trials (Sep 1) | 6.57% | 2.04x | High |
| 3 onboarding emails to steal (Sep 15) | 6.13% | 1.91x | High |
| 5 tools we cut (Aug 4) | 4.02% | 1.25x | No |
| Weekly roundup #31 to #34 | 2.07% to 2.42% | 0.64x to 0.75x | No (low, but above the 0.5x line) |
No send fell below 0.5x the median.
Shared traits of the high outliers: all three have a subject that names one specific subject (a pricing page, a named company's result, a set of onboarding emails), and two of three promise something concrete to take away ("teardown", "to steal"). None is a roundup.
| Trait | Testable? | Reason |
|---|---|---|
| Subject style | Yes | Subjects vary; see above |
| Topic or format (roundup vs single topic) | Yes | Inferred from subjects only |
| Send day | Not testable | Every send went out on a Tuesday, exactly a week apart |
| Length | Not testable | Not in the data (no send traits provided) |
| Lead story | Not testable | Not in the data (no send traits provided) |
Three findings with evidence
-
Single topic deep dives get roughly 2.6x the clicks of roundups, which matters directly for per click sponsor revenue. The four non roundup sends averaged 617.5 clicks (410, 720, 690, 650); the four roundups averaged 235 (250, 240, 230, 220). All three high outliers are single topic sends. This finding rests on 4 versus 4 sends, so it is a consistent pattern in this sample, not proof.
-
Roundups are declining every week, on both clicks and unsubscribes. Roundup clicks went 250, 240, 230, 220 while delivered grew from 10,350 to 10,650, so click rate fell from 2.42% to 2.07%. Over the same four roundups, unsubscribes rose 15, 18, 20, 22 (0.145% to 0.207%). The four highest unsubscribe rates in the whole period are the four roundups; the three lowest are the three high outliers. This is a steady trend across 4 sends, but it is still only 4 data points.
-
Subject wording may separate the deep dives: "5 tools we cut" (list, 4.02%) underperformed the other three single topic sends (6.13% to 6.99%). It is the only non roundup send below 1.5x the median. Possible reasons: a list of tools is closer to a roundup, or the subject is less specific. Insufficient data: this rests on one send, so treat it as a question to test, not a finding to act on.
Three tests
Test window: the next four sends.
| # | Hypothesis | What changes | Metric to watch |
|---|---|---|---|
| 1 | Replacing roundups with single topic deep dives raises clicks per send without raising unsubscribes. | Alternate: send 2 deep dives in place of 2 roundups, keep 2 roundups as the control. | Clicks per send and click rate on delivered; unsubscribe rate as the guardrail. |
| 2 | A roundup led by one featured deep dive item (with its own headline in the subject) recovers some lost clicks. | For the 2 remaining roundups, change the subject from "Weekly roundup #N" to the name of the lead item, and lead with that item. | Roundup click rate versus the 2.07% to 2.42% range of #31 to #34. |
| 3 | Sponsor placement near the specific, high interest content gets more sponsor clicks than placement elsewhere. | Place the sponsor block directly after the lead story in all 4 sends and send per link click counts afterward. | Sponsor link clicks per send (needs per link data, which I did not have for this review). |
Note: no live automation or account change is involved in this review, so no approval is needed. If you decide to stop the roundup format inside a scheduled or automated send, approve that change before it is made.